专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08942978B2 Parameter learning in a hidden trajectory model 有权
标题翻译：隐藏轨迹模型中的参数学习
公开(公告)号：US08942978B2
公开(公告)日：2015-01-27
申请号：US13182971
申请日：2011-07-14
申请人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
发明人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
IPC分类号： G10L15/00 , G10L15/06
CPC分类号： G10L15/063 , G10L2015/025
摘要： Parameters for distributions of a hidden trajectory model including means and variances are estimated using an acoustic likelihood function for observation vectors as an objection function for optimization. The estimation includes only acoustic data and not any intermediate estimate on hidden dynamic variables. Gradient ascent methods can be developed for optimizing the acoustic likelihood function.
摘要翻译：使用用于观察向量的声学似然函数作为优化的反对函数来估计包括装置和方差的隐藏轨迹模型的分布参数。该估计仅包括声学数据，而不包括对隐藏的动态变量的任何中间估计。可以开发梯度上升方法来优化声似然函数。

2. 发明授权

US08700552B2 Exploiting sparseness in training deep neural networks 有权
标题翻译：在深层神经网络训练中利用稀疏性
公开(公告)号：US08700552B2
公开(公告)日：2014-04-15
申请号：US13305741
申请日：2011-11-28
申请人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
发明人： Dong Yu , Li Deng , Frank Torsten Bernd Seide , Gang Li
IPC分类号： G06F15/18 , G06N3/08
CPC分类号： G06N3/08
摘要： Deep Neural Network (DNN) training technique embodiments are presented that train a DNN while exploiting the sparseness of non-zero hidden layer interconnection weight values. Generally, a fully connected DNN is initially trained by sweeping through a full training set a number of times. Then, for the most part, only the interconnections whose weight magnitudes exceed a minimum weight threshold are considered in further training. This minimum weight threshold can be established as a value that results in only a prescribed maximum number of interconnections being considered when setting interconnection weight values via an error back-propagation procedure during the training. It is noted that the continued DNN training tends to converge much faster than the initial training.
摘要翻译：提出了深层神经网络（DNN）训练技术实施例，其训练DNN，同时利用非零隐藏层互连权重值的稀疏性。通常，完全连接的DNN最初通过遍历完整的训练集多次进行训练。那么，在大多数情况下，只有重量大小超过最小重量阈值的互连在进一步的训练中被考虑。该最小权重阈值可以被建立为在训练期间通过错误反向传播过程设置互连权重值时仅考虑规定的最大数量的互连的值。值得注意的是，继续进行的DNN训练往往比初始训练快得多。

3. 发明申请

US20140067735A1 COMPUTER-IMPLEMENTED DEEP TENSOR NEURAL NETWORK 有权
标题翻译：计算机实现深度传感器神经网络
公开(公告)号：US20140067735A1
公开(公告)日：2014-03-06
申请号：US13597268
申请日：2012-08-29
申请人： Dong Yu , Li Deng , Frank Seide
发明人： Dong Yu , Li Deng , Frank Seide
IPC分类号： G06N3/08
CPC分类号： G06N3/02 , G06N3/04 , G06N3/0454 , G06N3/084
摘要： A deep tensor neural network (DTNN) is described herein, wherein the DTNN is suitable for employment in a computer-implemented recognition/classification system. Hidden layers in the DTNN comprise at least one projection layer, which includes a first subspace of hidden units and a second subspace of hidden units. The first subspace of hidden units receives a first nonlinear projection of input data to a projection layer and generates the first set of output data based at least in part thereon, and the second subspace of hidden units receives a second nonlinear projection of the input data to the projection layer and generates the second set of output data based at least in part thereon. A tensor layer, which can converted into a conventional layer of a DNN, generates the third set of output data based upon the first set of output data and the second set of output data.
摘要翻译：本文描述了深张量神经网络（DTNN），其中DTNN适合于在计算机实现的识别/分类系统中的使用。 DTNN中的隐藏层包括至少一个投影层，其包括隐藏单元的第一子空间和隐藏单元的第二子空间。隐藏单元的第一子空间至少部分地将输入数据的第一非线性投影接收到投影层，并且至少部分地生成第一组输出数据，并且隐藏单元的第二子空间接收输入数据的第二非线性投影投影层并且至少部分地基于其生成第二组输出数据。可以转换成DNN的常规层的张量层基于第一组输出数据和第二组输出数据产生第三组输出数据。

4. 发明授权

US08239195B2 Adapting a compressed model for use in speech recognition 有权
标题翻译：适应用于语音识别的压缩模型
公开(公告)号：US08239195B2
公开(公告)日：2012-08-07
申请号：US12235748
申请日：2008-09-23
申请人： Jinyu Li , Li Deng , Dong Yu , Jian Wu , Yifan Gong , Alejandro Acero
发明人： Jinyu Li , Li Deng , Dong Yu , Jian Wu , Yifan Gong , Alejandro Acero
IPC分类号： G10L15/20
CPC分类号： G10L15/20 , G10L15/065
摘要： A speech recognition system includes a receiver component that receives a distorted speech utterance. The speech recognition also includes an adaptor component that selectively adapts parameters of a compressed model used to recognize at least a portion of the distorted speech utterance, wherein the adaptor component selectively adapts the parameters of the compressed model based at least in part upon the received distorted speech utterance.
摘要翻译：语音识别系统包括接收失真的语音话语的接收机组件。所述语音识别还包括适配器组件，所述适配器组件选择性地适配用于识别所述失真语音话语的至少一部分的压缩模型的参数，其中所述适配器组件至少部分地基于接收失真的语音话语选择性地调整所述压缩模型的参数讲话话语。

5. 发明申请

US20110270610A1 PARAMETER LEARNING IN A HIDDEN TRAJECTORY MODEL 有权
标题翻译：参数学习在隐藏的TRAJECTORY模型
公开(公告)号：US20110270610A1
公开(公告)日：2011-11-03
申请号：US13182971
申请日：2011-07-14
申请人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
发明人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
IPC分类号： G10L15/00
CPC分类号： G10L15/063 , G10L2015/025
摘要： Parameters for distributions of a hidden trajectory model including means and variances are estimated using an acoustic likelihood function for observation vectors as an objection function for optimization. The estimation includes only acoustic data and not any intermediate estimate on hidden dynamic variables. Gradient ascent methods can be developed for optimizing the acoustic likelihood function.
摘要翻译：使用用于观察向量的声学似然函数作为优化的反对函数来估计包括装置和方差的隐藏轨迹模型的分布参数。该估计仅包括声学数据，而不包括对隐藏的动态变量的任何中间估计。可以开发梯度上升方法来优化声似然函数。

6. 发明授权

US08010356B2 Parameter learning in a hidden trajectory model 有权
标题翻译：隐藏轨迹模型中的参数学习
公开(公告)号：US08010356B2
公开(公告)日：2011-08-30
申请号：US11356898
申请日：2006-02-17
申请人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
发明人： Li Deng , Dong Yu , Xiaolong Li , Alejandro Acero
IPC分类号： G10L15/00
CPC分类号： G10L15/063 , G10L2015/025
摘要： Parameters for distributions of a hidden trajectory model including means and variances are estimated using an acoustic likelihood function for observation vectors as an objection function for optimization. The estimation includes only acoustic data and not any intermediate estimate on hidden dynamic variables. Gradient ascent methods can be developed for optimizing the acoustic likelihood function.
摘要翻译：使用用于观察向量的声学似然函数作为优化的反对函数来估计包括装置和方差的隐藏轨迹模型的分布参数。该估计仅包括声学数据，而不包括对隐藏的动态变量的任何中间估计。可以开发梯度上升方法来优化声似然函数。

7. 发明申请

US20100153104A1 Noise Suppressor for Robust Speech Recognition 有权
标题翻译：噪声抑制器用于强大的语音识别
公开(公告)号：US20100153104A1
公开(公告)日：2010-06-17
申请号：US12335558
申请日：2008-12-16
申请人： Dong Yu , Li Deng , Yifan Gong , Jian Wu , Alejandro Acero
发明人： Dong Yu , Li Deng , Yifan Gong , Jian Wu , Alejandro Acero
IPC分类号： G10L15/20
CPC分类号： G10L21/0208 , G10L15/20
摘要： Described is noise reduction technology generally for speech input in which a noise-suppression related gain value for the frame is determined based upon a noise level associated with that frame in addition to the signal to noise ratios (SNRs). In one implementation, a noise reduction mechanism is based upon minimum mean square error, Mel-frequency cepstra noise reduction technology. A high gain value (e.g., one) is set to accomplish little or no noise suppression when the noise level is below a threshold low level, and a low gain value set or computed to accomplish large noise suppression above a threshold high noise level. A noise-power dependent function, e.g., a log-linear interpolation, is used to compute the gain between the thresholds. Smoothing may be performed by modifying the gain value based upon a prior frame's gain value. Also described is learning parameters used in noise reduction via a step-adaptive discriminative learning algorithm.
摘要翻译：描述了通常用于语音输入的噪声降低技术，其中除了信噪比（SNR）之外，基于与该帧相关联的噪声电平来确定用于帧的噪声抑制相关增益值。在一个实现中，降噪机制基于最小均方误差，Mel-frequency cepstra降噪技术。设置高增益值（例如一个），以在噪声电平低于阈值低电平时实现很少或没有噪声抑制，以及设置或计算的低增益值，以实现高于阈值高噪声电平的大噪声抑制。使用噪声功率相关函数，例如对数线性插值来计算阈值之间的增益。可以通过基于先前帧的增益值修改增益值来执行平滑化。还描述了通过步进自适应识别学习算法在降噪中使用的学习参数。

8. 发明申请

US20100070279A1 PIECEWISE-BASED VARIABLE -PARAMETER HIDDEN MARKOV MODELS AND THE TRAINING THEREOF 有权
标题翻译：基于改进的可变参数隐藏式MARKOV模型及其训练
公开(公告)号：US20100070279A1
公开(公告)日：2010-03-18
申请号：US12211114
申请日：2008-09-16
申请人： Dong Yu , Li Deng , Yifan Gong , Alejandro Acero
发明人： Dong Yu , Li Deng , Yifan Gong , Alejandro Acero
IPC分类号： G10L15/14
CPC分类号： G10L15/144
摘要： A speech recognition system uses Gaussian mixture variable-parameter hidden Markov models (VPHMMs) to recognize speech under many different conditions. Each Gaussian mixture component of the VPHMMs is characterized by a mean parameter μ and a variance parameter Σ. Each of these Gaussian parameters varies as a function of at least one environmental conditioning parameter, such as, but not limited to, instantaneous signal-to-noise-ratio (SNR). The way in which a Gaussian parameter varies with the environmental conditioning parameter(s) can be approximated as a piecewise function, such as a cubic spline function. Further, the recognition system formulates the mean parameter μ and the variance parameter Σ of each Gaussian mixture component in an efficient form that accommodates the use of discriminative training and parameter sharing. Parameter sharing is carried out so that the otherwise very large number of parameters in the VPHMMs can be effectively reduced with practically feasible amounts of training data.
摘要翻译：语音识别系统使用高斯混合可变参数隐马尔可夫模型（VPHMM）来识别许多不同条件下的语音。 VPHMM的每个高斯混合分量的特征在于平均参数μ和方差参数＆Sgr。这些高斯参数中的每一个作为至少一个环境调节参数的函数而变化，例如但不限于瞬时信噪比（SNR）。高斯参数随环境条件参数变化的方式可以近似为分段函数，如三次样条函数。此外，识别系统制定均值参数μ和方差参数＆Sgr; 每个高斯混合分量以有效的形式适应使用歧视性训练和参数共享。执行参数共享，以便通过实际可行的训练数据量可以有效地减少VPHMM中非常大量的参数。

9. 发明授权

US07653535B2 Learning statistically characterized resonance targets in a hidden trajectory model 有权
标题翻译：在隐藏的轨迹模型中学习统计学上的共振目标
公开(公告)号：US07653535B2
公开(公告)日：2010-01-26
申请号：US11303899
申请日：2005-12-15
申请人： Li Deng , Dong Yu , Alejandro Acero
发明人： Li Deng , Dong Yu , Alejandro Acero
IPC分类号： G10L19/06 , G10L19/14 , G10L11/04 , G10L15/00
CPC分类号： G10L15/14 , G10L25/12 , G10L25/15 , G10L25/24
摘要： A statistical trajectory speech model is constructed where the targets for vocal tract resonances are represented as random vectors and where the mean vectors of the target distributions are estimated using a likelihood function for joint acoustic observation vectors. The target mean vectors can be estimated without formant data. To form the model, time-dependent filter parameter vectors based on time-dependent coarticulation parameters are constructed that are a function of the ordering and identity of the phones in the phone sequence in each speech utterance. The filter parameter vectors are also a function of the temporal extent of coarticulation and of the speaker's speaking effort.
摘要翻译：构建统计轨迹语音模型，其中声道共振的目标被表示为随机向量，并且使用关联声学观测向量的似然函数来估计目标分布的平均向量。可以不使用共振峰数据来估计目标平均向量。为了形成模型，构建了基于时间依赖的协方差参数的随时间依赖的滤波器参数矢量，其是每个语音话语中电话序列中的电话的排序和身份的函数。滤波器参数矢量也是协调的时间范围和说话者的说话力的函数。

10. 发明申请

US20090177468A1 SPEECH RECOGNITION WITH NON-LINEAR NOISE REDUCTION ON MEL-FREQUENCY CEPTRA 有权
标题翻译：语音识别与非线性噪声减少在频率CEPTRA
公开(公告)号：US20090177468A1
公开(公告)日：2009-07-09
申请号：US11970537
申请日：2008-01-08
申请人： Dong Yu , Alejandro Acero , James G. Droppo , Li Deng
发明人： Dong Yu , Alejandro Acero , James G. Droppo , Li Deng
IPC分类号： G10L15/20
CPC分类号： G10L15/20 , G10L15/02 , G10L21/02 , G10L25/24
摘要： In an automatic speech recognition system, a feature extractor extracts features from a speech signal, and speech is recognized by the automatic speech recognition system based on the extracted features. Noise reduction as part of the feature extractor is provided by feature enhancement in which feature-domain noise reduction in the form of Mel-frequency cepstra is provided based on the minimum means square error criterion. Specifically, the devised method takes into account the random phase between the clean speech and the mixing noise. The feature-domain noise reduction is performed in a dimension-wise fashion to the individual dimensions of the feature vectors input to the automatic speech recognition system, in order to perform environment-robust speech recognition.
摘要翻译：在自动语音识别系统中，特征提取器从语音信号中提取特征，并且基于提取的特征，通过自动语音识别系统识别语音。通过特征增强提供降噪作为特征提取器的一部分，其中基于最小均方误差准则提供了以Mel-frequency cepstra形式的特征域降噪。具体来说，设计的方法考虑了清洁语音和混合噪声之间的随机相位。为了执行环境鲁棒的语音识别，特征域噪声降低以维度方式执行到输入到自动语音识别系统的特征向量的各个维度。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式